Carbon sequestration by
Victorian inland wetlands
Carbon sequestration by Victorian inland wetlands
Carbon sequestration by Victorian inland
wetlands.
Citation
Carnell P, Windecker S, Brenker M, Yukate B, Johnson K and Macreadie P. 2016. Carbon
sequestration by Victorian inland wetlands. Blue Carbon Lab, Deakin University, Victoria, Australia.
ISBN 978-1-76047-220-7 (Print)
ISBN 978-1-76047-221-4 (pdf/online)
Acknowledgements
The authors would like to thank Kate Brunt from the Goulburn Broken CMA for guiding the project.
The broader steering committee including Janet Holmes (DELWP), Paul Reich (ARI), Rohan Hogan
(NCCMA), Natalie Dando (NECMA), Adam Bester (GHCMA), Simon Casanelia (GBCMA) and Tamara
van Polanen Petel (DELWP). A big thank you to wetland representatives from each CMA for
supplying wetland site information, coordinating with landholders and visiting us in the field while
sampling. We also thank Steve Krueger of Deakin University for his help with field monitoring, as
well as Jan Barton and other members of the Blue Carbon Lab including Tessa Evans, Quinn Olivier
and Alex Pearse.
Report produced by:
Blue Carbon Lab, Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin
University.
Disclaimer:
This publication may be of assistance to you, but Deakin University and its employees do not
guarantee that the publication is without flaw of any kind or is wholly appropriate for your particular
purposes and therefore disclaims all liability for any error, loss or other consequence which may
arise from you relying on any information in this publication.
2
Carbon sequestration by Victorian inland wetlands
List of Definitions
Carbon sequestration rate: The rate at which carbon dioxide and other forms of carbon is being
captured and stored within the soil. Expressed as mass of carbon per unit area per year (e.g. Mg C ha1
yr-1).
CO2 eq.: Stands for carbon dioxide equivalents. This is a standard unit for converting the amount of
carbon and methane into the equivalent amount of CO2.
CH4: Chemical symbol for methane.
Ha: Hectare (10,000 m2)
Mg: Mega-gram or metric tonne (1,000,000 grams).
Soil Accretion rate: The rate of increase in soil height due to sedimentation within a system, usually
expressed as millimetres per year (mm yr-1).
Soil Accumulation rate: The rate at which soil (both organic and mineral) is accumulating within a
system, generally expressed on a mass per unit area (e.g. Mg ha-1 yr-1) basis.
Soil: The upper layer of the earth, made up of a complex matrix of mineral and organic sediments of
different sizes (ranging from clay to boulders) from different sources (Mitsch and Gosselink 2007).
Soil carbon content: Amount of carbon (specified as either total, organic or inorganic) within a given
soil sample, also known as a carbon pool. Usually expressed as mass per unit area (e.g. g C cm -2).
Combining carbon content for all samples from a given soil core will give carbon stock of that core.
Soil carbon concentration: The concentration of carbon (specified as either total, organic or inorganic)
within the soil sample, generally expressed as grams per kilogram of soil (e.g. g C kg-1), or as a
percentage.
Soil inorganic carbon: Fraction of carbon within the soil which is mineral-based with the most common
form being calcium carbonate.
Soil organic carbon: Fraction of carbon within the soil which is derived from organic matter such a
plants.
Soil carbon stock: The amount of carbon stored within a system’s soils. Summation of the carbon
content values from a given core. Generally reported as either soil carbon stock (includes both
inorganic and organic fractions) or soil organic carbon (SOC) stock (only the organic fraction). Usually
presented either on a unit mass per area basis to a certain depth (e.g. Mg Corg ha-1 to 1 m soil depth)
or on just mass basis when estimating over large areal extents.
TC: Total Carbon. Includes both inorganic carbon (IC) and organic carbon (OC).
3
Carbon sequestration by Victorian inland wetlands
Table of Contents
Executive Summary................................................................................................................................. 5
Introduction ............................................................................................................................................ 7
Methods .................................................................................................................................................. 8
1.
Site selection ............................................................................................................................... 8
2.
Field sampling ............................................................................................................................. 8
3.
Laboratory Processing............................................................................................................... 10
4.
Carbon stock calculations ......................................................................................................... 11
5.
Carbon sequestration rate measurements ............................................................................... 12
6.
Estimated emissions from wetland losses since European settlement .................................... 12
7.
Data analysis ............................................................................................................................. 13
Results ................................................................................................................................................... 14
1.
Aim 1. Carbon stocks and sequestration rates across Victoria ................................................. 14
1.1. Carbon density, dry bulk density and % carbon values ............................................................ 14
1.2. Wetland comparisons ............................................................................................................... 15
1.3. Carbon sequestration rates ...................................................................................................... 17
1.4. Victoria wetland carbon stock and sequestration rate estimates ............................................ 19
2.
Aim 2. Carbon stock and sequestration hotspots ..................................................................... 20
3.
Aim 3. Estimated emissions from wetland losses since European settlement......................... 24
Discussion.............................................................................................................................................. 25
Assessing Victorian wetland carbon stocks and sequestration rates ........................................... 25
Identifying carbon sequestration hotspots ................................................................................... 27
Potential emissions from previous wetland degradation and loss ............................................... 29
Future research priorities ............................................................................................................. 29
Final remarks ................................................................................................................................. 31
References ............................................................................................................................................ 32
4
Carbon sequestration by Victorian inland wetlands
Executive Summary
Wetlands are areas of permanent, periodic, or intermittent inundation that hold still or very slow
moving water. Wetlands can be natural or human-made, and in Victoria include billabongs, lakes,
swamps, marshes, peatlands, floodplains, mangroves, mudflats, and seagrass areas. According to the
most recent (2013) inventory, the state’s 23,739 natural wetlands cover an area of 614,259 hectares,
and most (69%) occur on private land. Victoria also has a further 11,060 artificial wetlands spanning
170,613 hectares, plus 321 wetlands (2,702 ha) that have not been classified.
Victoria’s wetlands supply a range of values for ecosystem function and human well-being.
Traditional Owners of the land have long regarded wetlands as key areas for food and other
resources, making them an important part of the region’s cultural heritage. Within an agricultural
context, wetlands are used for stock grazing, water supply, and amenity. Wetlands also help trap
sediments, pollutants, and cycle nutrients: a particularly important function in coastal areas, in
which wetlands provide flood protection and help reduce the impacts of runoff. Victoria’s wetlands
are also important for supporting biodiversity (e.g. birds, fish, flora) and sustaining threatened
species (24% of threatened native species depend on wetlands for their survival).
Yet as the impacts of climate change are increasingly realised, wetlands are drawing more and more
attention for a different reason: they have vast potential for capturing atmospheric carbon. So far,
investigations of the carbon sequestration capacity of wetlands have concentrated on coastal or
‘blue carbon’ wetlands (i.e. seagrasses, saltmarshes, and mangroves). But in fact, estimates identify
inland wetlands as the earth’s largest store of terrestrial carbon: they contain 33% of the soil carbon
pool, yet occupy a mere 6–8% of the land surface. Despite this potentially extreme worth, wetlands
have historically been underappreciated; since European settlement, Victoria has seen widespread
losses of wetlands through agricultural development, urban development, and water extraction.
Crucially, wetland loss and degradation has two major consequences: the loss of carbon
sequestration capacity; and the potential release of ancient carbon back into the atmosphere – an
impact that effectively transforms wetlands from carbon sinks into carbon sources.
The purpose of this study was to survey Victoria’s inland wetlands (n = 103, sampled between
August 2015 and February 2016) for their carbon sequestration capacity. Wetlands were chosen
from a list of priority wetlands identified by each of the ten Victorian Catchment Management
Authorities (CMAs), and consisted of the following wetland categories: freshwater meadow, shallow
freshwater marsh, deep freshwater marsh, permanent open freshwater, semi-permanent saline
wetland, permanent saline wetland, and high country peatland. Sampling involved taking soil cores
(to 1 m; n = 3 cores per wetland) and analysing the total carbon content (via MIR and a CHN
analyser) at different depths (up to 6 samples) in each core. In total, we analysed 1,674 samples. To
determine carbon sequestration rates, we collated published data on soil accretion rates and utilised
the carbon stock information, to calculate soil carbon sequestration rates.
Carbon densities ranged from 0.0 to 260.87 mg Corg cm-3, with a mean + SEM (standard error of the
mean) of 31.45± 0.82 mg Corg cm-3. Carbon densities were generally highest in surface soils (0–2 cm),
dropping by approximately 50% by 48–50 cm. High country peatlands had the highest carbon stocks
(292.54 ± 177.60 Mg Corg ha-1); saline wetlands and permanent open freshwater had the lowest
carbon stocks (64.28 ± 47.60 and 113.02 ± 119.94 Mg Corg ha-1 respectively). Soil accretion rates
ranged from 1.02 to 30.00 mm yr-1, with a mean ± standard error of 6.66 ± 1.33 mm y-1. Mean
5
Carbon sequestration by Victorian inland wetlands
carbon sequestration rate for Victorian inland wetlands was estimated at 6.93 ± 1.37 Mg CO2 eq. ha-1
yr-1, with a total estimated carbon stock of 68 million Mg Corg.
The value of the carbon stock to depth of refusal of Victorian inland wetlands equates to $AUD
6 billion, and an annual carbon sequestration value of $AUD 76.74 million per year. The latter value
is based on annual sequestration of 3,117,682 tonnes of CO2 equivalents per year – equivalent to the
CO2 emissions of 176,538 Australians. Disturbance and loss of wetlands also has the potential to
release significant quantities of CO2 back into the environment. Here, we estimated that since
European settlement, such losses have released 22.5–74.2 million Mg CO2 equivalents (based on
27.25 to 90% of soil carbon lost). Loss of permanent open freshwater wetlands accounts for the
largest proportion (43%) of the estimated emissions, while deep freshwater marshes and shallow
freshwater marshes account for 22.7% and 22.5% respectively.
In summary, we recommend the following actions be taken to maximize wetland carbon stocks in
Victoria:
1) Prioritize wetland carbon hotspots for conservation and protection from disturbance.
2) Focus revegetation projects in areas with higher sediment accretion rates, such as deep or
shallow freshwater marshes and riverine wetlands.
3) Restore natural hydrology (e.g. via bund wall/drain removals or through environmental
water delivery) to enhance how wetlands sequester carbon.
Overall, via the most comprehensive investigation of inland wetland carbon stocks in Australia, this
study confirms that Victoria’s inland wetlands represent significant carbon sinks. As Australia seeks
to capitalise on new biosequestration opportunities, there is great interest in restoring such habitats
for carbon offset purposes; indeed, the federal government announced that it will be among the first
countries to include wetlands in its National Greenhouse Gas Inventory.
However, to facilitate this move, research must also now examine how inland wetlands might
contribute greenhouse gases, because as well as capturing CO2, these systems are also the world’s
largest source of methane – a potent greenhouse gas. In Australia, there are currently too few data
to develop fully-constrained carbon budgets that balance the carbon sequestration capacity of
inland wetlands with their release of greenhouse gases. Therefore, we recommend future studies
that:
1) measure greenhouse gas fluxes (particularly methane) from inland wetlands;
2) determine the opportunity for carbon offsetting in wetlands in Victoria;
3) determine how carbon stocks and greenhouse gas release is affected by environmental
and land-use change (e.g. grazing, wet-dry cycles); and
4) examine whether it is possible to restore wetlands in ways that minimise methane release
and maximise carbon sequestration.
6
Carbon sequestration by Victorian inland wetlands
Introduction
Wetlands are some of the most productive habitats in the world. They support a unique biodiversity
(Strayer and Dudgeon, 2010), and provide important ecosystem services, such as nutrient cycling,
erosion control, and flood mitigation (Mitra et al. 2005, Mitsch and Gosselink 2007, de Groot et al.
2012, Costanza et al. 2014, Russi et al. 2013). But rising carbon dioxide levels in the atmosphere have
brought attention to another important service of wetland systems: the capacity for wetlands to
store carbon. By taking up carbon during photosynthesis and growth and burying undecomposed
plant litter in anaerobic soils, wetlands store an average of 34 to 44 times more carbon than do
terrestrial forests (McCleod et al. 2011, Bernal and Mitsch 2012). Global estimates suggest that
freshwater wetlands contain 33% of the soil carbon pool: a disproportionately high contribution
given that they occupy a mere 6–8% of the land surface (Mitra et al. 2005, Lal 2007, Mitsch and
Gosselink 2007).
Despite their importance, wetlands have been historically underappreciated, and an estimated 87%
of global wetland area has been lost since 1700 (Davidson 2014). They are threatened by pollution,
transformation, water extraction, and modification (MEA 2005, Moser et al. 1996, van Asselen et al.
2013, Vörösmarty et al. 2010). Such disturbance can undermine a wetland’s ability to capture
carbon, but critically, damage to wetlands can also release significant amounts of carbon that have
already been stored (Lal et al. 2007, Page and Dalal 2011, Pendleton et al. 2012). Conversion to
agricultural land for cropping and grazing, in particular, has negative consequences on wetland soil
carbon storage and sequestration. For example, Sigua et al. 2009 measured a 96% reduction in soil
organic carbon (%) when a wetland was converted to beef cattle pasture, while Meyer et al. (2008),
estimated an 80% loss of soil carbon after roughly 40 years of soybean and corn cultivation.
Quantifying the contribution of wetland ecosystems to carbon capture and storage is vital not only
to ensuring these valuable systems are managed to maximise storage, but to providing extra
justification for protection of areas valuable for a multitude of other services.
While research suggests that inland freshwater wetlands may provide significant carbon storage,
there is high variability across this diverse range of systems making generalisation across
hydrogeomorphic types difficult. For example, in Ohio, USA, depressional wetlands sequester 2.25
times more carbon than riverine communities (317 ± 93 gC m-2 yr-1 compared to 140 ± 16 gC m-2 yr-1;
Bernal and Mitsch 2012). Within these broader hydrogeomorphic categories, different plant
communities (Mitsch et al 2013, Villa and Mitsch 2015) also impact carbon sequestration rates.
There are few studies that attempt to systematically compare carbon stocks and sequestration rates
between wetland types and across catchments within a single region. Specifically, this project aims
to: 1) undertake the first broad-scale assessment of carbon stocks of inland Victorian wetlands; 2)
identify carbon hotspots based on these data; and 3) assess the potential impact of historical
wetland loss on carbon stocks.
7
Carbon sequestration by Victorian inland wetlands
Methods
This study was conducted across the state of Victoria in southeastern Australia (Figure 1). The inland
catchments of Victoria contain 530,413 hectares of wetlands, but an estimated 201,175 ha (or
roughly 26%) have been lost since European settlement (Papas and Moloney 2012). The wetlands
that still exist in Victoria are also vary in condition. In a recent state-wide assessment of close to 600
wetlands in Victoria, almost two thirds of wetlands on public land were in good or excellent
condition, compared to 39% on private land (DSE 2012). Moderate to very poor hydrology was found
in 46% of wetlands, and 19% showed a moderate, poor or very poor assessment for disturbance to
soils (DSE 2012). Given the broadscale nature of the survey, our study sampled mostly wetlands in
good condition across both public and private land to produce a best case scenario estimate of the
carbon value of our remaining wetland resources.
1. Site selection
To evenly spread the sampling effort across the state, we sampled at least 10 wetlands in each
Catchment Management Authority (CMA) region, totalling 103 wetlands (Figure 1). Wetlands were
chosen from a list of priority wetlands identified by each CMA, but also across six key wetland
categories defined by Corrick and Norman (1980): freshwater meadow, shallow freshwater marsh,
deep freshwater marsh, permanent open freshwater, permanent saline wetland, and high country
peatland.
2. Field sampling
Between August 2015 and February 2016 we collected five soil cores at each of 103 wetlands across
the ten participating CMA regions. At each site we selected a single dominant vegetation stratum to
sample within, to ensure soil was collected from areas with the same overlying litter material as well
as similar inundation level. Cores were taken 50 m apart from one another in an attempt to sample
more widely across the sites.
At each of the five sampling points, a 5 cm (inner-diameter) PVC pipe (the core) was hammered into
the soil until 1 m was reached, or until core refusal (no further penetration). Soil compaction was
calculated based on the depth of soil contained in the core compared to the depth of the core in the
ground. These values are then later used to correct the depth of the sample taken (eg. with
compaction of 0.75, a 15 cm sample becomes 18.75 cm). To ensure the soil remained stable during
removal, a rubber plug was inserted into the top of the core to create a vacuum seal. After removal,
a foam plug was inserted down the top of the pipe to maintain the sediment stratigraphy during
transport. Both ends of the core were capped and duct taped to prevent moisture loss. All cores
were brought back to the laboratory at Deakin University in Burwood, Victoria, for subsequent
processing.
At each sampling point, a 1 m2 quadrat centred on the collection site of each core was used to aid in
visual estimation of species cover and height. These data were collated to verify wetland type and
dominant vegetation classifications obtained from online spatial databases (DELWP Biodiversity
Interactive Map at www.depi.vic.gov.au/environment-and-wildlife/biodiversity/biodiversityinteractive-map and the Wetland Current spatial layer at www.data.vic.gov.au).
8
Figure 1. Extant wetlands and waterbodies in Victoria and the location of the 103 wetlands sampled in this study.
3. Laboratory Processing
In the laboratory, soil from three of the five replicate cores from each site were extruded and
sectioned at 0–2 cm, 12–14 cm, 28–30 cm, 48—50 cm, 74—76 cm, and 98—100 cm for carbon stock
calculations (Figure 2). Any dead plant material was left in the sections and the sections were then
dried until consistent weight at 50oC for 48-72 hours. Dry weight was used to calculate sediment
bulk density (weight (g) / volume (cm3)) and then homogenized with a stainless steel mortar and
pestle (Retch RM200).
0 -2 cm
14 - 16 cm 28 - 30 cm
48 - 50 cm
74 -76 cm
Figure 2. Example sediment core illustrating the depth sections sampled in this study (core taken
from Seaford Wetland in Port Phillip and Western Port CMA).
At 11 of the 103 sites, the fourth core was processed for carbon analysis and 210Pb age dating, and
was sectioned every centimetre from 0–20 cm. The sections for 210Pb dating were split in half, with
one half to be age dated, and the other reserved for carbon analysis, so that carbon sequestration
values could be calculated.
As some of the samples were collected from declared Phylloxera Infested Zones (areas infected with
an aphid-like insect pest of grapevines (Daktulosphaira vitifoliae), all cores were gamma irradiated at
50kGray for 60 hours (by Steritech, Dandenong, Victoria), before being sent to the soil carbon lab at
CSIRO, South Australia, for analysis. This irradiation process has been demonstrated to have no
major influence on carbon measurements in soil samples (Balldock et al. unpublished data).
Based on the protocols of Baldock et al. (2013), a Thermo Nicolet 6700 FTIR spectrometer equipped
with a Pike AutoDiff automated diffuse reflectance accessory was used to obtain diffuse reflectance
Fourier-transform MIR spectra for all samples across a spectral range of 8700-400 cm-1 at 8 cm-1.
Using Unscrambler X ver 10.1 software, a principle component analysis (PCA) was used to visualize
spectra variability and the Kennard-Stone Algorithm (Kennard & Stone 1969) was used to pick the
most representative 286 samples from the total dataset. The resultant model was then used to
predict Corg values for the remaining 1386 samples.
For the 198 samples, total carbon (TC) and total nitrogen (TN) were determined in the laboratory.
High temperature (1350°C) oxidative combustion on a LECO Trumac CN analyser was used to
estimate TC and TN, using lance oxygen flows and an extended purge to ensure complete
combustion of carbonates. Non-calcareous samples identified based on MIR spectra (absence of
reflectance peak at 2500cm-1) had no further analysis (TC = TOC). Here we present the Corg results.
Carbon sequestration by Victorian inland wetlands
4. Carbon stock calculations
The depth of refusal at each site varied (some as shallow as 16 cm), but most cores reached at least
50 cm. We followed the approach for calculating total sedimentary carbon stock as outlined by
Howard et al. (2014). We calculated soil carbon density by multiplying the dry bulk density by the
percent carbon (Step 1). To obtain carbon stocks for the full 2 cm section of sediment, we multiplied
the Corg density (mg Corg cm-3) calculated for each depth interval by two (Step 2). For each carbon
stock core with more than three depth data points, cubic splines were used to calculate the Corg
values for intermediate depth ranges that were not analysed (Step 3). Although cubic splines are
commonly preferred for soil carbon analysis (e.g. Minasny et al. 2006), linear splines are more
appropriate for cores where there were only two or three data points per core. Carbon stock values
for all depth intervals (both measured and estimated) were summed to calculate carbon stocks (per
cm2) to the depth sampled for each core (Step 4). This value was used to upscale Corg stocks to Mg
Corg ha-1 using the following unit conversion factors: 1,000,000 g = 1 Mg (megagram), and
100,000,000 cm2 = 1 hectare.
Step 1.
Soil carbon density (g/cm3) = dry bulk density (g/cm3) * (% Corg /100)
Step 2.
Carbon content in core sections (g/cm3) = Soil carbon density (g/cm3) * Thickness of core section (2
cm)
Step 3.
Create cubic equations (splines) for each core, using the depth samples measured, to estimate
carbon density in each 2 cm section throughout the entire core.
Step 4.
Sum the carbon density values for each 2 cm section, to get a total carbon stock for the entire core.
Step 5.
Total sedimentary carbon (MgC ha-1) = Averaged core carbon (g/cm3) * (1 Mg/1,000,000 g)
*(100,000,000 cm2/1 hectare)
11
Carbon sequestration by Victorian inland wetlands
5. Carbon sequestration rate measurements
The profiles of 210Pb concentrations were determined for one sediment core at each 11 of the 103
wetlands. Analysis of its decay product 210Po, in equilibrium with 210Pb was used to determine
concentrations of 210Pb (Sanchez-Cabeza, Masque & Ani-Ragolta 1998). Soil samples were spiked
with known amounts of 209Po, acid digested, plated and their emissions were measured by alpha
spectrometry using Passivated Implanted Planar Silicon (PIPS) detectors (CANBERRA, Mod. PD450.18 A.M). The concentrations of 210Pb were calculated by applying an appropriate decay–
ingrowth correction and accounting for blank (one per 10 samples) and detector background, which
were both almost negligible (awaiting final results). Analyses of samples and reference materials
were carried out in parallel.
Concentrations of 226Ra (Radium isotope) were determined by gamma spectrometry using a highpurity Germanium (Ge) well-type detector (CANBERRA, mod. GCW3523) through the 295 and 351
keV emission lines of 214Pb. Samples were stored in sealed containers for 3 weeks prior to counting
to attain equilibrium between 226Ra and its short-lived decay products. Concentrations of 210Pbex
were obtained by subtracting the 210Pbsup from the total 210Pb at each section. The concentrations of
226
Ra were in agreement with those of 210Pb in the deepest sections of the cores, where no excess
210
Pb (210Pbex) was present. All sediment cores had similar concentrations of supported 210Pb (i.e. in
equilibrium with 226Ra; 210Pbsup = awaiting final results). The model of constant rate of supply (CRS;
Appleby & Oldfield 1978), which assumes a constant flux of 210Pbex to the sediment surface, was used
to date the sediment based on 210Pbex inventories and estimate sediment accretion rates in the
cores. The CRS model was adapted for those sites with missing inventories of 210Pbex, following
Appleby (2001).
Beyond carbon sequestration measurements taken in this study, there have also been a number of
other studies that have calculated sediment accretion rates in wetlands in Victoria (Gell et al. 2009,
Gell et al. 2014). We compiled these data sets, and combined the sediment accretion rate (mm yr-1)
with carbon density estimates from up to three wetlands from the current stock assessment (Table
2). Carbon density values from stock assessment sites allowed estimation of the average carbon
sequestration rate at the literature sites.
6. Estimated emissions from wetland losses since European settlement
Carbon emissions from wetland loss were calculated based on carbon stock estimates from this
study in combination with data on pre-European (pre-1778s) and present-day distribution. Area of
ecosystem loss from each CMA was calculated by comparing pre-European wetland extent layer
with current wetland extent. Next, average carbon density of CMA, calculated by averaging
measurements from 1-5 wetlands that occurred within each sector, were multiplied by the area of
loss to estimate total sediment carbon stocks that were previously present. Where sediment cores
had not been collected within the CMA, the average carbon stock value for the CMA was based on
the average of that wetland type in all of Victoria.
The upper, lower, and intermediate estimates of carbon remineralization were calculated based on
values used in previous studies. Murray et al. (2011) estimated that 90% of the OC in the top meter
of sediment is released as CO2 emissions, while Donato et al. (2011) estimated that 50% of the top
30 cm would be converted and 25% from the remainder of the top meter of sediment. Siikamaki et
al. (2013) combined these estimates to calculate a high (90%), low (27.25%) and intermediate
12
Carbon sequestration by Victorian inland wetlands
(58.625%) estimate of the carbon stock lost as CO2 equivalents (eq.). The emissions estimates are
presented here as 90, 58.625, and 27.25% remineralization of Corg from to the depth that was
sampled in this study, converted from Corg to CO2 eq. by multiplying by 3.67 (Pendleton et al. 2012).
7. Data analysis
Carbon stock values to refusal were transformed (log10(Mg Corg ha-1)) prior to statistical analysis to
meet assumptions of normality and equal variances. Carbon stock was compared across ecosystems
using a one-way ANOVA, with carbon stock as the response and wetland type as the factor. A posthoc Tukey Pairwise Comparison was performed to distinguish differences in population means. A
fully nested ANOVA was performed to compare carbon stock in each ecosystem across the six CMA
regions where they co-occurred, using carbon stock as the response and wetland type and CMA as
factors.
Sediment sample extruded in the lab.
13
Carbon sequestration by Victorian inland wetlands
Results
We have set up the results section to reflect information relating to the aims and objectives of this
project. Section 1) undertakes the first broad-scale assessment of carbon stocks of inland Victorian
wetlands, 2) identifies carbon hotspots based on these data, and 3) assesses the potential impact of
historical wetland loss on carbon stocks.
1. Aim 1. Carbon stocks and sequestration rates across Victoria
1.1. Carbon density, dry bulk density and % carbon values
The organic carbon density of all samples collected from wetlands across Victoria, ranged from 0.0
to 260.87 mg Corg cm-3, with a mean + SEM (standard error of the mean) of 31.45± 0.82 mg Corg cm-3.
Supplementary Table 1 provides information on how carbon density varied with depth. Carbon
density was generally highest at the surface (0—2 cm) and at 14—16 cm, with a 25% decline by 28—
30 cm and 50% by 48—50 cm. However, there were a number of sites that did not follow this trend.
The Dry Bulk Density of all samples ranged from 0.03 to 2.35 g cm-3, with an average of 1.01 ± 0.02 g
cm-3 (Figure 3a). This is a relatively high dry bulk density value compared to other wetland studies
(Meyer et al. 2008, Whitacker et al. 2015). The percentage Corg values across all samples ranged from
0.0 to 55.85 %, with an average of 7.72 ± 0.31% (Figure 3b).
a)
b)
Figure 3. (a) Organic carbon content (% carbon, n=1251) and (b) dry bulk density (DBD, g, cm-3,
n=1251) samples across the 103 wetland sampled, plotted by depth sampled corrected for sediment
compaction.
14
Carbon sequestration by Victorian inland wetlands
1.2. Wetland comparisons
Average surface (0 – 2 cm) soil total carbon (Corg) density varied greatly between wetland types,
24.03 ± 20.86 (permanent open freshwater), 35.07 ± 21.04 (freshwater meadow), 41.15 ± 25.68
(shallow freshwater marsh) and 46.68 ± 35.11 (deep freshwater marsh), 50.94 ± 24.35 (alpine
peatlands) and 22.76 ± 14.32 (saline wetlands) mg Corg cm-3 (Table 3). When comparing soil carbon
stocks across the five wetland types most sampled, permanent open freshwater wetlands had
significantly lower carbon stocks compared to shallow freshwater marsh, deep freshwater marsh
and alpine peatlands ((ANOVA, F2, 857=140.20, p <0.001; Figure 3, Supplementary table 1). This
equated to average sediment carbon stocks of 113.02 ± 119.94 Mg Corg ha-1 for permanent open
freshwater, 201.26 ± 196.49 Mg Corg ha-1 for shallow freshwater marsh, 233.31 ± 192.20 Mg Corg ha-1
for deep freshwater marsh, and 292.54 ± 177.60 Mg Corg ha-1 for high country peatlands. Alpine
peatlands also had higher carbon stocks when compared to freshwater meadows 126.41 ± 104.19
Mg Corg ha-1 (Supplementary Table 1).
Carbon stocks varied significantly between CMA regions, but these differences also depended on the
wetland type. For deep freshwater marshes, Glenelg Hopkins had the highest carbon stocks with an
average of 376.03 ± 269.42 Mg Corg ha-1 (Table 1, Supplementary Table 1). Port Phillip and Western
Port had the highest average carbon stocks for shallow freshwater marsh, with an average of 426.63
± 194.21Mg Corg ha-1 (Table 1, Supplementary Table 1). In comparison, for alpine peatlands, there
was no significant difference between CMAs.
Figure 4. Average organic carbon stocks to refusal (n= 317 cores) for the various wetland types,
according to the Corrick and Norman (1980) classification.
15
Carbon sequestration by Victorian inland wetlands
Table 1. Overview of the wetland types sampled within each CMA region and the average (± SD) organic carbon stock (Mg Corg ha-1)
Saline
Corangamite
East Gippsland
Freshwater
Shallow
Deep freshwater
High country
freshwater
meadow
freshwater marsh
marsh
peatlands
250.21 ±147.09
287.34 ± 54.11
263.13 ± 177.91
314.40 ± 247.99
177.94 ± 0.52
234.32 ± 42.69
Glenelg Hopkins
127.04 ± 88.53
18.64 ± 6.24
Goulburn Broken
75.01 ± 41.92
Mallee
78.08 ± 28.05
North Central
73.52 ± 51.87
Permanent open
20.79 ± 3.59
North East
175.24 ± 53.48
491.30 ± 279.31
112.97 ± 25.24
132.32 ± 53.10
58.48 ± 13.56
7.97 ± 3.78
88.99 ± 65.55
25.66 ± 7.08
122.36 ± 97.35
130.45 ± 19.14
86.85 ± 32.99
123.10 ± 48.78
418.41 ± 174.18
508.98 ± 55.92
319.54 ± 287.82
275.13 ± 138.59
Port Phillip & Western Port
115.66 ± 20.59
179.03 ± 108.57
West Gippsland
Wimmera
61.57 ± 17.83
37.14 ± 25.18
42.23 ± 17.12
66.11 ± 35.77
94.42 ± 57.58
State Average
64.28 ± 47.60
113.02 ± 119.94
126.41 ± 104.19
201.26 ± 196.49
233.31 ± 192.20
16
386.94 ± 281.42
367.34± 105.64
232.32 ± 116.24
292.54 ± 177.60
Carbon sequestration by Victorian inland wetlands
1.3. Carbon sequestration rates
Sediment accretion rates were compiled from a number of published studies in southeast Australian
wetlands (Table 2). Soil accretion rate can be translated into carbon sequestration rate by
multiplying by carbon density. Soil accretion rates ranged from 1.02 to 30.00 mm yr-1, with a mean ±
standard error of 6.66 ± 1.33 mm y-1. Using carbon density data from our carbon measurements,
average carbon sequestration rates were 6.93 ± 1.37 Mg CO2 eq. ha-1 yr-1.
Previous syntheses of sediment accretion rates by Gell et al. (2009) indicate a substantial increase in
sediment accretion rate since European settlement. This is evident in the high values of sites located
along the River Murray, including King’s Billabong, Pikes creek, Barmah, Hogans 1, and Callemondah
1. This increase in sediment accretion has presumably also resulted in increased carbon
sequestration in these wetlands.
a)
b)
Figure 5. a) Barmah Lake in Barmah National Park: one of the sites with the highest carbon
sequestration rates in Victoria. b) Lake Colac, the site with the lowest carbon sequestration rate.
17
Carbon sequestration by Victorian inland wetlands
Table 2. Estimated carbon sequestration rates in wetlands, utilising other age dating studies in southeastern Australia. Using this information on sediment
accretion rate (SAR), we then choose the closest wetlands of a similar type, and averaged these to estimate the Corg density of the surface sediment (0—2
cm), and used this to calculate the carbon sequestration rate in CO2 equivalents, per hectare, per year.
Site
CMA
Reference
Latitude
(S)
Kings Billabong
Junction Park Billabong
Hopcrofts Billabong
Lake Cullulleraine
Tareena Billabong
Ral Ral Creek
Pikes Creek
Willsmere Billabong
Bolin Billabong
Delegate River
Barmah Forest
Hogans 1
no. 9
no. 11
no. 19
no. 23
no. 32
no. 38
Callemondah 1
Callemondah 2
Lake Curlip
Lake Surprise
Lake Elingamite
Lake Colac
Lake Purrumbete
Wetland average + SEM
Mallee
NSW
Mallee
Mallee
NSW
SA
SA
PPW
PPW
EG
GB
NSW
NSW
NE
NSW
NE
GB
GB
GB
GB
EG
GH
GH
C
C
Kattell et al 2015
Gell et al 2009
Gell et al 2009
Gell et al 2009
Gell et al 2005b
Gell et al 2009
Gell et al 2009
Lintern et al 2016
Lintern et al 2016
Gell Stuart Smith 1993
Thoms et al 1999
Reid et al 2007
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Thoms et al 1999
Ladd 1978
Barr et al 2014
Barr et al 2014
Gell et al 2012
Tibby et al 2012
-34.239
-34.772
-34.715
-34.273
-34.061
-34.045
-34.295
-37.783
-37.769
-37.229
-35.862
-36.025
-35.927
-36.129
-36.025
-36.133
-36.008
-35.870
-36.695
-36.689
-37.750
-38.061
-38.356
-38.309
-38.284
Longitude
(E)
SAR
(mm yr-1)
Estimated Corg
density 0-2cm
(mg Corg cm-3)
Estimated Corg
sequestration rate
(Mg CO2 eq. ha-1 yr-1 )
Estimated carbon
sequestration value
($ha-1 yr-1)
142.226
143.301
143.151
141.597
141.234
140.744
140.656
145.056
145.078
148.839
145.322
146.715
147.713
146.953
146.227
146.210
145.809
145.372
145.160
145.164
148.565
141.923
143.004
143.589
143.230
14.29
3.80
5.00
5.00
3.50
9.50
30.00
9.40
4.00
3.47
21.80
10.00
7.37
2.53
4.10
3.85
5.30
10.00
1.70
3.20
1.40
4.24
5.93
1.02
2.46
6.66 ± 1.33
28.31
27.41
27.41
17.15
27.41
27.41
27.41
32.79
32.79
36.44
27.48
19.69
19.69
24.39
19.69
23.70
19.69
27.48
23.00
23.00
20.40
52.33
37.12
5.91
41.09
29.68 ± 2.58
14.84
3.82
5.03
3.15
3.52
9.56
30.18
11.31
4.81
4.64
21.99
7.23
5.33
2.26
2.96
3.35
3.83
10.09
1.44
2.70
1.05
8.14
8.08
0.22
3.71
6.93 ± 1.37
358.5
92.3
121.5
76.0
85.0
230.8
728.9
273.2
116.2
112.1
531.0
174.5
128.6
54.7
71.6
80.9
92.5
243.6
34.7
65.2
25.3
196.7
195.1
5.3
89.6
167.35 + 32.97
18
Carbon sequestration by Victorian inland wetlands
1.4. Victoria wetland carbon stock and sequestration rate estimates
Using the carbon stock data from the sites sampled, Victoria’s total wetland soil organic carbon
stocks are estimated to be over 68 million Mg Corg (Table 3). Given that not all sites could be sampled
to the full 1 m depth, this stock value is likely to be an underestimate as the soil in most cases would
have extended beyond the depth that we could sample using this technique. Permanent open
freshwater wetlands are responsible for 18% of Victoria’s wetland carbon stock, mostly due to their
extensive area in Victoria (about 30% of the wetland area). Freshwater meadows store 16% of
Victoria’s inland wetland soil carbon stocks, while saline wetlands only account for 9% of soil carbon
stocks despite accounting for almost 25% of the state’s wetlands by area. Conversely, despite their
high soil carbon stocks, due to their limited distribution (0.73% of total inland wetland area), alpine
peatlands store 1.14 % of inland wetland soil carbon stocks in Victoria.
By converting the estimated carbon stocks for Victoria to CO2 equivalents and using a 2013-14
carbon price of $AUD 24.15/tonne, Victoria’s carbon stocks are valued at $AUD 6 billion. Similarly, by
using the average carbon sequestration rate from the wetlands listed in Table 2 (6.93 ± 1.37 Mg CO2
eq. ha-1 yr-1), to the total area of freshwater wetlands in Victoria (458,540 ha), the estimated carbon
sequestration of Victoria’s freshwater wetlands is 3,177,682 tonnes of CO2 equivalents per year. This
has a potential value of $AUD 76.74 million per year, or is equivalent to the CO2 emissions of
176,538 Australians.
Table 3. Wetland sediment carbon density, average total organic carbon stock, total wetland area in
Victoria and our estimated organic carbon stocks for Victoria by wetland type (Corrick and Norman
1980). These values are not extrapolated to a standard measure; cores are sampled in the field to
refusal and are therefore site specific. These values are combined with the total area of each
wetland type, to produce an estimate of organic carbon stock by wetland type.
Number
Average of 0 – 2 cm
Average of Mg
Total
Total stocks
of cores
mg Corg cm-3 ± SD
Corg ha-1 ± SD
Victorian Area
(Mg Corg)
Deep freshwater marsh
100
46.68 ± 35.11
233.31 ± 192.20
(Hectares)
55,790.10
15,004,315.98
Freshwater meadow
32
35.07 ± 21.04
126.41 ± 104.19
144,179.50
14,199,789.49
High country peatlands
36
50.94 ± 24.35
292.54 ± 177.60
4,475.70
190,355.78
Permanent open freshwater
67
24.03 ± 20.86
113.02 ± 119.94
185,034.90
19,825,164.75
Saline
21
22.76 ± 14.32
64.28 ± 47.60
155,718.50
10,005,798.46
Shallow freshwater marsh
57
41.15 ± 25.68
201.26 ± 196.49
69,060.70
8,891,555.50
Grand Total
316
38.50 ± 28.65
185.74 ± 175.90
614,259.40
68,116,979.96
Wetland type
19
Carbon sequestration by Victorian inland wetlands
2. Aim 2. Carbon stock and sequestration hotspots
The wetlands with the highest carbon stock values are listed in Table 4. Here, McKenzie’s Rd bog in
West Gippsland CMA has the highest average carbon stock, followed by Long swamp in Glenelg
Hopkins CMA and Highland peatland site 4 in Goulburn Broken CMA. While these wetlands have the
highest carbon stock values per hectare, the wetlands with the largest estimated carbon stocks are
Lake Corangamite, Lake Coleman, Heart Morass, Long Swamp and Ewings Morass, due to their large
area.
Carbon Hotspot mapping (Figure 6) visually overlays the average soil organic carbon content from
the wetland types sampled and within each CMA region (Table 1).Where we did not have data for a
wetland type in a CMA, we used the state average value. It provides a method for identifying high
value habitat patches using spatial information. The current analysis highlights that the wetlands
along the Murray River, southwestern Victoria and in the Gippsland Lakes region are some of the
highest value blue carbon locations (Figure 7; a and b).
Mckenzie’s Rd Bog, West Gippsland CMA.
20
Carbon sequestration by Victorian inland wetlands
Table 4. The top twenty wetlands, listed in order of highest average Organic carbon stock. The estimated wetland organic carbon stock is calculated from
the average organic carbon stock to refusal multiplied by the current wetland area. The estimated organic carbon stock value of this wetland is then
calculated from the estimated wetland organic carbon stock, by converting to CO2 equivalents and using the 2014 carbon price of $24.15 per Mg of CO2
equivalents.
Site
CMA
Wetland type
Area (Ha)
Average organic carbon
stock (Mg Corg ha-1)
McKenzie's Rd Bog
Long Swamp
Highlands Peatlands 4
Lang Lang Swamp
Bryans Swamp
Rhyll Swamp
Delegate River Bog
Lake Purrumbete
Boneo Swamp
Heart Morass
Nunziata Bog
Seaford Wetlands
Dereel Lagoon
Cooper's Bog
Cabbage Tree Lagoon
Bunyip Reserve
Dam Valley Bog
Lake Curlip
Lake Elingamite
Lara Swamp
WG
GH
GB
PPW
GH
GH
EG
C
PPW
WG
NE
GH
C
M
EG
PPW
WG
EG
GH
WG
Shallow freshwater marsh
Deep freshwater marsh
High country peatlands
Shallow freshwater marsh
Deep freshwater marsh
Deep freshwater marsh
High country peatlands
Permanent open
Shallow freshwater marsh
freshwater
Deep freshwater marsh
High country peatlands
Shallow freshwater marsh
Deep freshwater marsh
High country peatlands
Deep freshwater marsh
Freshwater meadow
High country peatlands
Deep freshwater marsh
Permanent open
Deep freshwater marsh
freshwater
2
726.2
4.6
5.6
666.3
43.1
176
527.8
320.2
1560.3
4
93.2
33.6
3.8
442.9
39.1
1.3
944.5
278.9
6
780.31 ± 24.38
631.68 ± 382.31
585.43 ± 262.77
575.75 ± 99.38
550.53 ± 114.78
508.98 ± 55.92
507.97 ± 168.42
489.56 ± 159.04
473.32 ± 107.83
402.13 ± 102.67
397.10 ± 86.11
391.00 ± 104.07
352.03 ± 51.04
337.58 ± 133.52
332.19 ± 51.27
331.20 ± 64.06
315.26 ± 136.10
301.54 ± 72.02
289.58 ± 53.24
284.47 ± 247.61
21
Estimated wetland organic
carbon stock (Mg Corg)
Estimated Organic
Carbon stock value
($)
1,560.63
458,725.17
2,692.96
3,224.20
366,815.71
21,936.85
89,402.38
258,388.94
151,557.67
627,440.76
1,588.39
36,441.25
11,828.17
1,282.80
147,125.99
12,949.85
409.83
284,808.71
80,763.18
1,706.80
138,319
40,657,041
238,678
285,762
32,511,060
1,944,274
7,923,778
22,901,141
13,432,632
55,610,388
140,779
3,229,807
1,048,337
113,696
13,039,850
1,147,752
36,324
25,242,739
7,158,081
151,274
Carbon sequestration by Victorian inland wetlands
Figure 6. Wetland carbon hotspots in Victoria. Higher carbon stock values (to depth of refusal) are represented by red, and lower values by green
22
Carbon sequestration by Victorian inland wetlands
a)
b)
Figure 7. Wetland carbon hotspots in Victoria. Higher carbon stock values (to depth of refusal) are
represented by red, and lower values by green a) The central Murray River region, which includes
Goulburn Broken CMA and North Central CMA. b) Southwestern Victoria with Glenelg Hopkins CMA
and Corangamite CMA’s.
23
Carbon sequestration by Victorian inland wetlands
3. Aim 3. Estimated emissions from wetland losses since European settlement
Based on our carbon stock measurements, wetland ecosystems present at the time of European
settlement that have since been lost had an estimated total carbon stock of over 82.5 million Mg
CO2 equivalents (Table 5). Forty-three percent of carbon stocks in these lost ecosystems were
associated with permanent open freshwater wetland loss, while deep freshwater marshes and
shallow freshwater marsh account for 22.7% and 22.5 % of carbon stocks, respectively. Looking at
the impacts of disturbance down to the top meter of sediments, conservative, intermediate, and
high estimates of emissions range between 22.5 million to 74.2 million Mg CO2 equivalent (Table 5).
Largest estimated carbon stock losses from deep freshwater marshes occurred in West Gippsland,
with substantial losses of permanent open freshwater in Corangamite, totalling to almost 40% of the
emissions resulting from wetland loss.
Table 5. Estimated sediment carbon emissions based on ecosystem loss since European settlement.
CO2 equivalent losses were calculated based off averages for each region, however for simplicity
here this is presented by wetland type across all CMAs.
CO2 Equivalent loss
Area Lost
(Ha)
Stock
Mg Corg Ha-1 Avg.
27.25%
58.63%
90%
13,312
233
5,112,099
10,998,987
16,883,998
Freshwater meadow
24,749
126
2162666
4653104
7142749
Permanent open freshwater
71,525
113
9,752,089
20,982,202
32,208,735
Saline
5,872
64
381,500
820,819
1,259,998
Shallow freshwater marsh
31,595
201
5,059,179
10,885,126
16,709,216
State Total
260,530
281
22,467,533
48,340,238
74,204,696
Sites
Deep freshwater marsh
24
Carbon sequestration by Victorian inland wetlands
Discussion
This report provides the most comprehensive assessment of carbon stocks and sequestration rates
of inland wetlands in Australia to date. Victoria’s wetlands are storing substantial amounts of
carbon, with a total carbon stock of 68 million Mg Corg estimated for the depths sampled. This
equates to a potential value of $AUD 6 billion. And the estimated carbon sequestration of Victoria’s
freshwater wetlands is 3,117,682 tonnes of CO2 equivalents per year. This has a value of $AUD 76.74
million per year, equivalent to the CO2 emissions of 176,538 Australians. At the same time,
disturbance and loss of wetlands has the potential to release significant quantities of CO2 back into
the environment. Since European settlement, the loss of wetlands has released an estimated 22.5
million to 74.2 million Mg CO2 equivalents, depending on the emission factor applied.
Assessing Victorian wetland carbon stocks and sequestration rates
There was significant variation in carbon stocks between wetland types and across CMA regions.
Carbon stock accumulation typically depends on: 1) availability of carbon supply (whether externallyderived carbon travels into the wetland due to its low geomorphic setting in the landscape, or
whether carbon is internally-derived from the deposits of plant material); 2) quality of carbon supply
(less complex carbon forms are more easily decomposed, and therefore do not remain long in the
soil profile); and 3) the conditions available to the decomposer community (the environment
conditions both impact the rate of decomposition directly, and select for different microbial
communities). All of these factors can contribute to explaining patterns observed at the study sites.
Permanent open freshwater wetlands (typically freshwater lakes with minimal fringing vegetation)
had relatively low average carbon stocks compared to the other wetland types (Figure 4, Table 1).
This is not surprising given the ecology of these systems. While permanent freshwater lakes can be
sinks for catchment carbon, their hydrology means they may be relatively isolated from major
waterways (such as the Murray), and fed only by small-scale runoff from their own catchment. In
addition, as they are typically too deep to sustain emergent vegetation, there is little internally
produced carbon to sequester. While these water bodies are important to the landscape for their
other functions, our results appear to confirm other research that suggests these ecosystems are not
particularly important for blue carbon sequestration.
Inland saline waterbodies are an interesting addition to our study. As indicated in Table 3, they make
up a large proportion of Victoria’s inland waterbodies, and are therefore critical to understand.
While the carbon stocks are not particularly high, this is partly because most saline wetlands could
not be sampled past 30 cm depth. If these stocks were to be extrapolated to 50 or 100 cm, this value
would be significantly higher. While the estimates here suggest that, because of their large area,
saline wetlands make important contributions to the whole stock of the state, it is important to note
that all present measurements were made in vegetated areas, and it is unclear how carbon stocks
and sequestration rates may differ in vegetated and unvegetated areas. Thus, further work is
needed to clarify how the unvegetated components of inland saline wetlands (e.g. Lake
Corangamite) contribute to carbon storage. While we have no data on sequestration rates in saline
wetlands, rates are likely to be high due to the reduced decomposition fostered by more saline
conditions. As saline areas also inhibit methanogenesis, such waterbodies also present an important
comparison for future research into the greenhouse gas emissions of freshwater wetlands.
25
Carbon sequestration by Victorian inland wetlands
Freshwater meadows and shallow freshwater marshes were on average in the mid-range for
wetland carbon stocks across the state. These wetlands are some of the most variable because they
encompass such a large range of inundation regimes. While it is convenient to use these wetland
categories in this study, future work should attempt to tease apart the relative contribution of
hydrology (inundation source, frequency, interval, and volume, among other factors), and factors
such as climate, geomorphology (elevation on the landscape), and vegetation (though the latter is
often tied to hydrologic regime). Inundation regime impacts soil conditions contributing to
decomposition rates, as well as the overlying carbon supply (aboveground vegetation). In
permanently inundated sites, anaerobic conditions prevail, which reduces the rate of decomposition
and enhances the storage of carbon. It is as yet unclear how varying wetting and drying regimes
impact carbon storage rates, but this will be another important avenue for future research on this
class of wetlands. In addition, periods of prolonged drying may result in loss of entire vegetation
communities for periods of time.
Deep freshwater marshes contain large carbon stock, as well as cover a large area across the state.
For this reason, these areas represent a key frontier for inland wetland carbon reserves. Deep
freshwater marshes, while variable, tend to have several characteristics in common. These areas are
flooded for a relatively long proportion of the year, and sustain dense vegetation year-round.
Despite these commonalities, we still found the highest variation within this class. One explanation
for this is that the Corrick and Norman classification scheme is not completely accurate across the
state. The integrity of future state-wide research will rely on continuing to improve the state-wide
wetland databases.
As expected, our study found the highest carbon stock values in alpine peatlands. Peat is terrestrial
soil composed of at least 20% organic matter. Peatlands occur in areas with higher water inflow than
outflow, or where there is a large supply of groundwater. Waterlogged soil produces the reduced,
anaerobic conditions that inhibit activity of microbial decomposers, allowing carbon-rich,
undecomposed plant matter to accumulate. Accumulation of plant detritus into peat can produce
dams, which further modify hydrology to favour continuing peat formation. While it is not surprising
that these systems had the highest stock, the results usefully validate the analysis methods used in
this report. Alpine peatlands, while contributing the smallest area in Victoria, are ecologically critical
carbon stocks, and also some of the most vulnerable to loss.
In this study, wetlands contained carbon stocks equivalent to (and often higher than) those reported
for other freshwater wetlands in Australia (Webb 2002, Page and Dalal 2011, Whitaker et al. 2015).
For the sub-tropical wetlands of Queensland, Bryant et al. (2008) reported average total carbon (%)
values from 0.25 % to 25.45% in the top 10 cm of soil, with large variation between wetland types
and regions. In comparison, the Victorian wetlands sampled show an even wider range of values,
from 0.0007 to 52.31 %, with an average stock of 7.86 ± 0.31%. The Queensland study also
encompassed wetlands across the state, which suggests that Victoria’s diversity of climatic regions
(semi-arid, temperate, coastal and alpine) and wetland types creates substantial variation.
Understanding the factors driving this variability in carbon stocks is an important next step in the
analysis of the data collected in this study.
26
Carbon sequestration by Victorian inland wetlands
Identifying carbon sequestration hotspots
Carbon stocks varied markedly across the ten Victorian regions, but all wetlands sampled represent
valuable carbon sinks, providing both ecological and economic benefits (Table 4, Figure 7).
McKenzie’s Rd had the highest carbon stocks per hectare. Located within Won Wron state forest,
McKenzie’s Rd Swamp is relatively pristine, with limited vehicle access. While Long Swamp in Glenelg
Hopkins CMA is more remote, the mixture of pine plantations to the north and a history of modified
hydrology, highlights areas for improved management of this important wetland. Fortunately, a
number of wetland rehabilitation efforts have recently taken place in Long Swamp to restore the
natural hydrology. Tracking the improvement in carbon stocks with wetland restoration such as this
is an important next step for wetland carbon research.
Lang Lang swamp is located near to what used to be the Koo Wee Rup swamp, which was drained
since European settlement. This small wetland, is probably indicative of the wetlands that made up
the Koo Wee Rup swamp complex. Boneo Swamp is also precariously placed in the landscape. Boneo
Swamp is surrounded by urban development, golf courses, a sewage treatment plant, waste transfer
station, is bisected by roads, and has a modified hydrology (Chinaman’s Creek, which feeds the
swamp, has been channelized).
Notably, the highest carbon sequestration values were often associated with sites located along
some of the major rivers in Victoria (or closer to fluvial inputs), such as the Murray, Goulburn, and
Yarra (Table 2). This is contrary to a study in Ohio, which found higher carbon sequestration rates in
isolated, depressional wetlands (Bernal and Mitsch 2012). However, the age dating studies in
Victoria have been predominantly conducted along riverine billabongs, so it is difficult to determine
if this is a real pattern, or part of sampling bias. Additional data on carbon sequestration rates in
isolated, depressional wetlands is needed to determine this pattern.
The carbon hotspot maps we provide are a useful tool to help managers easily identify wetlands
with high carbon stocks. This will enable managers to prioritise, for protection or rehabilitation,
those wetlands at greatest risk of disturbance through development, agriculture or altered
hydrology. The carbon hotspot maps can also be overlain with GIS layers of other ecosystem values
such as biodiversity and flood mitigation, to identify areas that may be protected for multiple
ecosystem services. Assessing the opportunities for both protection and rehabilitation of wetlands,
and the likely carbon offset value of such moves, is a vital next step if we are to move forward with
carbon offset opportunities in Victorian inland wetlands (Figure 8).
27
Carbon sequestration by Victorian inland wetlands
a)
b)
c)
d)
e)
f)
Figure 8. a) Cattle grazing in wetlands can cause significant loss of blue carbon through b) pugging
and disturbance of the soil. Large feral herbivores can also have impacts on wetlands and carbon,
including c) horses. This can occur through d) the grazing activity on the wetland plants, and through
soil disturbance via trampling. Wetland restoration activities can involve restoring natural
hydrological flow. This may involve e) filling drains that were put in by early Europeans to keep
water in the wetland (Image from http://natureglenelg.org.au/), and f) installing culverts to reconnect water flow that may have been disrupted by roads or other human structures.
28
Carbon sequestration by Victorian inland wetlands
Potential emissions from previous wetland degradation and loss
We estimate a loss of 22.5–74.2 million Mg CO2 eq., or 0.022–0.074 Pg CO2 eq., due to loss of
wetlands since European settlement in Victoria. This is equivalent to the annual emissions of
1,248,196 – 4,122,483 people or 4,780,326 – 15,788,234 cars. It is important to note that these
values are estimates only, and are based on current values reported in the international literature
for how wetland loss influences carbon stocks. These values are quite high, given that loss of
Indonesian mangroves has been estimated to have released 0.175 Pg CO2 eq. into the atmosphere
(Murdiyaso et al 2015). The high emissions from wetland loss in Victoria highlights the need to
ensure sufficient protection of wetlands to avoid ongoing carbon emissions.
One way to guard against continuing carbon losses is by protecting wetlands that may be used for
grazing or cropping. The impact of such activities are stark. For example, Sigua et al. (2009)
measured a 96% reduction in soil organic carbon (%) when a wetland was converted to beef cattle
pasture, while Meyer et al. (2008) estimated an 80% loss of soil carbon after roughly 40 years of
soybean and corn cultivation. This conversion to agriculture likely promoted soil moisture
fluctuations, which then stimulated decomposition and re-release of carbon (Sigua et al. 2009).
Further research is needed to confirm the impact of conversion on Victorian inland wetlands in a
controlled study.
It is here that wetland rehabilitation works may help to restore wetlands’ carbon sequestration
capacity. A recent study in the Murray Local Land Services region of New South Wales has
demonstrated how wetland rehabilitation activities can increase the soil organic carbon stock:
carbon stocks increased with time since protection (Carnell et al. 2016b). And globally, studies have
shown that wetland rehabilitation increases the carbon storage capacity of previously degraded
wetlands (Badiou et al. 2011). In a study of rehabilitated wetlands in the Canadian prairie pothole
region, Badiou et al. (2011) estimated soil organic carbon stocks to 30 cm at 121, 165, and 205 Mg
organic carbon ha-1 for newly re-habilitated (1–3 years), long-term rehabilitated (5–12 years), and
reference wetlands, respectively.
This study and others (Meyer et al. 2008, Ballantine and Schneider 2009) suggests that we can make
meaningful carbon offsets in inland wetlands through protection and rehabilitation measures (Erwin
2009). Various wetland rehabilitation projects are currently occurring across the state of Victoria
(Bachman and Holland 2015, Goulburn Broken CMA pers. comm.), and in future, measuring the
carbon offset benefit of these activities will provide important insights. With a number of methods
for measuring carbon gains through management activities recently developed for wetlands (VCS
2014, Carnell et al. 2016a), management agencies are now well placed to undertake wetland
protection and rehabilitation activities for the purpose of carbon offsetting.
Future research priorities
Below we outline the five key research steps to developing and implementing carbon offset
opportunities in Victorian wetlands (Figure 9). The soil carbon measurements in this study represent
one half of the net carbon budget for freshwater wetlands: while freshwater wetlands are important
soil carbon stores, they are also the largest natural source of methane. Thus, to account for release
of greenhouse gases from wetlands (namely methane (CH4) and nitrous oxide (NO2)), seasonal or
ideally, continuous measurements of these greenhouse gases wetlands, across wet and dry cycles (a
1–3 year period) would enable us to build a more complete picture of carbon dioxide and methane
29
Carbon sequestration by Victorian inland wetlands
dynamics in these wetlands (similar to Badiou et al. 2011). These data could be combined with those
collected here on carbon stock and sequestration rates, to calculate the net carbon budget for such
wetlands. This step should be a priority for ongoing work (Table 6 ).
To move forward with carbon offset initiatives in wetlands, we also need to better understand how
wetland restoration activities influence carbon sequestration. While some preliminary work on
carbon stocks in rehabilitated wetlands in the Murray Local Land Services region in NSW is promising
(Carnell et al. 2016b), it is only through monitoring carbon sequestration rates before and after
wetland rehabilitation that we can calculate the carbon offset from that activity. Carnell et al.
(2016a) have developed a wetland carbon monitoring manual to track increases in carbon
sequestration following wetland rehabilitation activities. The priority now is to implement these
recommendations across a range of rehabilitated wetlands.
Broadscale soil carbon stocks & sequestration rates
Broadscale
assessment of
carbon stocks
and
sequestration
rates
Greenhouse Gas emmisions
Restoration
Greenhouse
Gas emmisions
from wetlands? How do
wetland
rehabilitation
activities
increase carbon
sequestration
capacity?
Degradation
How does
wetland
degradation
influence
carbon stocks
and greenhouse
gas emmisions?
Opportunities
What
opportunities
exist for
conducting
carbon
offsetting using
inland
wetlands?
Figure 9. Steps for developing knowledge on wetlands’ capacity to store carbon, and how different
management activities can promote higher carbon storage. Steps in green have been completed,
steps in yellow are soon to get underway (or have been completed in nearby areas), and steps in
orange still need to be undertaken.
30
Carbon sequestration by Victorian inland wetlands
Table 6. Research portfolio questions and nominal ranking.
Research Questions
Funded?
If not, rank
Q 1. What is the opportunity in Victoria to offset carbon emissions by restoration or
protection of wetlands?
1
Q 2. How do different wetting and drying cycles in wetlands influence greenhouse gas
emissions from wetlands?

Focus of recently successful ARC Linkage (LP160100061) with Blue Carbon
Lab/Deakin University, Southern Cross University and Murray LLS in the semiarid wetlands of Murray LLS region in NSW.

Katy Limpert from the Blue Carbon Lab will also be working on similar
questions in Wimmera and North Central CMA’s in Victoria for her PhD.
Q 3. Are particular plant communities or plant traits associated with higher carbon
gains?

Current focus of a University of Melbourne and Blue Carbon Lab PhD project
by Saras Windecker.
Funded in
particular
regions, not
Victoria-wide
Mostly
(Holsworth
Wildlife
Research
Endowment)
Q 4. How do different wetland rehabilitation interventions influence soil accretion
(carbon sequestration) rates over time?
2
Q 5. What is the effect of long-term protection and rehabilitation of wetlands on
methane and carbon dioxide release?
3
Q 6. Does the inundation regime influence the response of wetlands to rehabilitation
(ie. do more frequently wet sites increase in carbon sequestration capacity faster?)
4
Q 7. What is the effect (short-term and long-term) of livestock trampling on soil
carbon stock and greenhouse release from wetlands?
5
Final remarks
This study provides the largest and most comprehensive carbon stock and sequestration rate
measurements for any region in Australia. This is an important step to developing wetland carbon
projects regionally, and given a growing carbon offset market, provides an added incentive for
rehabilitating and protecting wetlands. With the Australian Federal Government recently
announcing their intention to include wetlands in the National Greenhouse Gas Inventory, research
into the carbon offset capacity of wetlands will continue to fill an important scientific and applied
research knowledge gap.
31
Carbon sequestration by Victorian inland wetlands
References
Appleby PG and Oldfield F (1978) The calculation of lead-210 dates assuming a constant rate of
supply of unsupported 210Pb to the sediment. Catena 5:1-8
Appleby PG (2001) Chronostratigraphic techniques in recent sediments. In Last, W.M. and Smol, J.P.,
editors. Tracking environmental change using lake sediments 1:171–203
Bachmann M and Holland E (2015) Protecting and Restoring Iconic Wetlands in Australia. National
Wetlands Newsletter, Published by the Environmental Law Institute. 37(1).
Badiou P, McDougal R, Pennock D and Clark B (2011) Greenhouse gas emissions and carbon
sequestration potential in restored wetlands of the Canadian prairie pothole region. Wetlands
Ecology and Management 19:237-256. doi: 10.1007/s11273-011-9214-6
Baldock JA, Hawke B, Sanderman J and Macdonald LM (2013) Predicting contents of carbon and its
component fractions in Australian soils from diffuse reflectance mid-infrared spectra. Soil Research
51:577-595.
Barr C, Tibby J, Gell P, Tyler J, Zawadzki A and Jacobsen G (2014) Climatic variability in southeastern
Australia over the last 1500 years inferred from the fossil diatom records of two crater lakes.
Quaternary Science Reviews 95: 115-131.
Bernal B and Mitsch WJ (2012) Comparing carbon sequestration in temperate freshwater wetland
communities. Global Change Biology 18:1636-1647. doi: 10.1111/j.1365-2486.2011.02619.x
Bryant KL, Wilson PR, Biggs AJW, Brough DM and Burgess JW (2008) Soil Indicators of Queensland
Wetlands: Statewide assessment and methodology. Department of Natural Resources and Water,
Brisbane.
Carnell P, Barton J, Lester R and Macreadie P (2016a). Wetland carbon monitoring program manual.
Report to Murray LLS. Blue Carbon Lab, Deakin University, Victoria, Australia.
Carnell P, Barton J, Lester R, Brenker M and Macreadie P (2016b). Rehabilitating wetlands and their
carbon sequestration capacity. Report to Murray LLS. Blue Carbon Lab, Deakin University, Victoria,
Australia.
Corrick AH and Norman FI (1980) Wetlands of Victoria I. Wetlands and waterbirds of the Snowy River
and Gippsland Lakes catchment. Royal Society of Victoria 91:1-15.
Costanza R, de Groot R, Sutton P, van der Ploeg S, Anderson SJ, Kubiszewski I, Farber S and Turn RK
(2014) Changes in the global value of ecosystem services. Global Environmental Change 26:152-158
32
Carbon sequestration by Victorian inland wetlands
Davidson NC (2014) How much wetland has the world lost? Long-term and recent trends in global
wetland area. Marine and Freshwater Research 65:934-941.
Davy MC and Koen TB (2013) Variations in soil organic carbon for two soil types and six land uses in
the Murray Catchment, New South Wales, Australia. Soil Research 51:631-644.
de Groot R, Brander L, van der Ploeg S, Costanza R, Bernard F, Braat L, Christie M, Crossman N,
Ghermandi A, Hein L, Hussain S, Kumar P, McVittie A, Portela R, Rodriguez LC, ten Brink P and van
Beukering P (2012) Global estimates of the value of ecosystems and their services in monetary units.
Ecosystem Service 1:50–61.
Gell PA, Stuart IM and Smith JD (1993) The response of vegetation to changing fire regimes and
human activity in East Gippsland, Victoria, Australia. The Holocene 3:150-160.
Gell P, Bulpin S, Wallbrink P, Bickford S and Hancock G (2005) Tareena Billabong - A
palaeolimnological history of an ever changing wetland, Chowilla Floodplain, lower Murray-Darling
Basin. Marine and Freshwater Research 56: 441-456.
Gell P, Fluin J, Tibby J, Hancock G, Harrison J, Zawadzki A, Haynes D, Khanum S, Little F and Walsh B
(2009) Anthropogenic Acceleration of Sediment Accretion in Lowland Floodplain Wetlands, MurrayDarling Basin, Australia. Geomorphology 108: 122-126.
Gell P, Mills K and Grundell R (2012) A legacy of climate and catchment change: the real challenge
for wetland management. Climate change and Australian wetlands. Hydrobiologia. doi:
10.1007/s10750-012-1163-4.
Gell P and Reid M (2014) Assessing change in floodplain wetland condition in the Murray Darling
Basin. The Anthropocene, doi:10.1016/j.ancene.2014.12.002.
Kattell G, Gell P, Perga ME, Jeppesen E, Grundell R, Weller S, Zawadzki A and Barry L (2015) Tracking
a century of change in trophic structure and dynamics in a floodplain wetland: integrating
palaeoecological and palaeoisotopic evidence. Freshwater Biology 60:711-723.
Kennard RW and Stone LA (1969) Computer aided design of experiments. Technometrics 11:137148.
Ladd PG (1978) Vegetation History at Lake Curlip in Lowland Eastern Victoria, from 5200 B.P. to
Present. Australian Journal of Botany 26:393-414.
Lal R, Follett RF, Stewart BA and Kimble JM (2007) Soil Carbon Sequestration to Mitigate Climate
Change and Advance Food Security. Soil Science 172:943-956.
33
Carbon sequestration by Victorian inland wetlands
Lintern A, Leahy PJ, Zawadzki A, Gadd P, Heijnis H, Jacobsen G, Connor S, Deletic A and McCarthy DT
(2016) Sediment cores as archives of historical changes in floodplain lake hydrology. Science of The
Total Environment 544: 1008-1019.
Mcleod E, Chmura GL, Bouillon S, Salm R, Bjork M, Duarte CM, Lovelock CE, Schlesinger WH and
Silliman BR (2011) A blueprint for blue carbon: toward an improved understanding of the role of
vegetated coastal habitats in sequestering CO2. Frontiers in Ecology and the Environment 9:552-560.
doi: 10.1890/110004.
Meyer CK, Baer SG and Whiles MR (2008) Ecosystem Recovery Across a Chronosequence of Restored
Wetlands in the Platte River Valley. Ecosystems 11:193-208. doi: 10.1007/s10021-007-9115-y.
Minasny B, McBratney AB, Mendoca-Santos ML, Odeh IOA and Guyon B (2006) Prediction and digital
mapping of soil carbon storage in the Lower Namoi Valley. Australian Journal of Soil Research
44:233-244.
Mitra S, Wassmann R and Vlek PL (2005) An appraisal of global wetland area and its organic carbon
stock. Current Science 88:25.
Mitsch WJ, Bernal B, Nahlik AM, Mander U, Zhang L, Anderson CJ, Jorgensen SE and Brix H (2013)
Wetlands, carbon, and climate change. Landscape Ecology 28:583-597. doi: 10.1007/s10980-0129758-8.
Mitsch WJ and Gosselink JG (2007) Wetlands. Hoboken. John Wiley & Sons, Inc.
Moser M, Prentice C and Frazier S (1996) A global overview of wetland loss and degradation. In
Proceedings of the 6th Meeting of the Conference of Contracting Parties 10:19-27.
Murdiyarso D, Purbopuspito J, Kauffman JB, Warren MW, Sasmito SG, Donato DC, Manuri S,
Krisnawati H, Taberima and Kurnianto S (2015) The potential of Indonesian mangrove forests for
global climate change mitigation. Nature Climate Change 5: 1089-1092.
Page K and Dalal R (2011) Contribution of natural and drained wetland systems to carbon stocks,
CO2, N2O, and CH4 fluxes: an Australian perspective. Soil Research 49:377-388.
Papas P and Moloney P (2012) Victoria’s wetlands 2009-2011: statewide assessments and condition
modelling. Arthur Rylah Institute for Environmental Research 229.
Pendleton L, Donato DC, Murray BC, Crooks S, Jenkins WA, Sifleet S, Craft C, Fourqurean JW,
Kauffman JB, Marba N, Megonigal P, Pidgeon E, Herr D, Gordon D and Baldera A (2012) Estimating
34
Carbon sequestration by Victorian inland wetlands
global “blue carbon” emissions from conversion and degradation of vegetated coastal ecosystems.
Plos One 7:1-7.
Reid MA, Sayer CD, Kershaw AP and Heijnis H (2007) Palaeolimnological evidence for submerged
plant loss in a floodplain lake associated with accelerated catchment erosion (Murray River,
Australia). Journal of Paleolimnology 38:191-208.
Russi D, ten Brink P, Farmer A, Badura T, Coates D, Förster J, Kumar R and Davidson N (2013) The
economics of ecosystems and biodiversity for water and wetlands. IEEP, London and Brussels.
Sanchez-Cabeza J, Masque P and Ani-Ragolta I (1998) 210Pb and210Po analysis in sediments and
soils by microwave acid digestion. Journal of Radioanalytical and Nuclear Chemistry 227: 19-22.
Sigua GC, Coleman SW and Albano J (2009) Beef cattle pasture to wetland reconversion: Impact on
soil organic carbon and phosphorus dynamics. Ecological Engineering 35:1231-1236. doi:
10.1016/j.ecoleng.2009.05.004.
Strayer DL and Dudgeon D (2010) Freshwater biodiversity conservation: recent progress and future
challenges. Journal of the North American Benthological Society 29:344-358.
Tibby J, Penny D, Leahy P and Kershaw AP (2012) Vegetation and water quality responses to
Holocene climate variability in Lake Purrumbete, Western Victoria. Terra Australis 43: 359-373.
Thoms MC, Ogden RW and Reid MA (1999) Establishing the condition of lowland floodplain rivers: a
palaeo-ecological approach. Freshwater Biology 41:407-423.
Van Asselen S, Verburg PH, Vermaat JE and Janse JH (2013) Drivers of wetland conversion: A global
meta-analysis. Plos One 8:e81292.
Villa JA and Mitsch WJ (2015) Carbon sequestration in different wetland plant communities in the
Big Cypress Swamp region of southwest Florida. International Journal of Biodiversity Science,
Ecosystem Services & Management 11:17-28.
Vörösmarty CJ , McIntyre PB, Gessner MO, Dudgeon D and Prusevich A (2010) Global threats to
human water security and river biodiversity. Nature 467:555-561.
Whitaker K, Rogers K, Saintilan N, Mazumder D, Wen L and Morrison RJ (2015) Vegetation
persistence and carbon storage: Implications for environmental water management for Phragmites
australis. Water Resources Research 51:5284-5300.
35